Uploaded December 2024 | Updated September 2026, 1 week ago
Tune into our webinar series, SIAM MPE Community Meetings, organized by the SIAM Activity Group on Mathematics of Planet Earth: siam.org/membership/activity-groups/detail/mathematics-of-planet-earth.
This presentation was led by Pedram Hassanzadeh, University of Chicago, who discussed the AI revolution in weather/climate modeling and the challenges with interpretability and predicting gray swan weather extremes.
This video explores the growing role of deep neural networks (NNs) in improving weather forecasting and climate modeling, focusing on both short-term predictions and longer-term climate change studies. We highlight successes in data-driven models and hybrid weather-climate approaches, as well as the challenges in understanding their learning processes and predicting rare extreme weather events. The presentation also discusses potential solutions, including Fourier analysis and rare-event sampling, to overcome these hurdles and enhance AI's application to climate science.
#webinarseries #WeatherModeling #ClimateModeling #atmosphericscience #weatherpredictions #climatescience #meteorology #environmentalscience #ScientificResearch #weatherforecast #climatechange #ScienceWebinar #FourierAnalysis
Tune into our webinar series, SIAM MPE Community Meetings, organized by the SIAM Activity Group on Mathematics of Planet Earth: siam.org/membership/activity-groups/detail/mathematics-of-planet-earth.
This presentation was led by Pedram Hassanzadeh, University of Chicago, who discussed the AI revolution in weather/climate modeling and the challenges with interpretability and predicting gray swan weather extremes.
This video explores the growing role of deep neural networks (NNs) in improving weather forecasting and climate modeling, focusing on both short-term predictions and longer-term climate change studies. We highlight successes in data-driven models and hybrid weather-climate approaches, as well as the challenges in understanding their learning processes and predicting rare extreme weather events. The presentation also discusses potential solutions, including Fourier analysis and rare-event sampling, to overcome these hurdles and enhance AI's application to climate science.
#webinarseries #WeatherModeling #ClimateModeling #atmosphericscience #weatherpredictions #climatescience #meteorology #environmentalscience #ScientificResearch #weatherforecast #climatechange #ScienceWebinar #FourierAnalysis
